In recent years, dielectric barrier discharge ionization detectors (which are hereinafter abbreviated as the “BIDs”) employing the ionization by dielectric barrier discharge plasma have been put to practical use as a new type of detector for GC (for example, see Patent Literatures 1 and 2 as well as Non Patent Literature 1).
BIDs described in the aforementioned documents are roughly composed of a discharging section and a charge-collecting section which is located below the discharging section. In the discharging section, a low-frequency AC high voltage is applied to plasma-generating electrodes circumferentially formed around a tube made of a dielectric material, such as quartz glass (“dielectric tube”), to ionize an inert gas supplied into the tube line of the dielectric tube and thereby form atmospheric-pressure non-equilibrium plasma. Due to the effects of the light emitted from this plasma (vacuum ultraviolet light), excited species and other elements, the sample components in a sample gas introduced into the charge-collecting section are ionized. The resulting ions are collected through a collecting electrode to generate detection signals corresponding to the amount of ions, i.e. the amount of sample components.
FIG. 8 shows the configuration of the discharging section and surrounding area in the aforementioned BID. As noted earlier, the discharging section 610 includes a cylindrical dielectric tube 611 made of a dielectric material, such as quartz, the inner space of which forms a passage of inert gas serving as plasma generation gas. On the outer wall surface of the cylindrical dielectric tube 611, three ring-shaped metallic electrodes (made of stainless steel, copper or the like) are circumferentially formed at predetermined intervals of space. A high AC excitation voltage power source 615 for generating a low-frequency high AC voltage is connected to the central electrode 612 among the three electrodes, while the electrodes 613 and 614 located above and below the central electrode are both grounded. Hereinafter, the central electrode is called the “high-voltage electrode” 612, while the upper and lower electrodes are respectively called the “ground electrodes” 613 and 614. The three electrodes are collectively referred to as the plasma generation electrodes. Since the wall surface of the cylindrical dielectric tube 611 is present between the passage of the inert gas and the plasma generation electrodes, the dielectric wall itself functions as a dielectric coating layer which covers the surface of those electrodes 612, 613 and 614, enabling a dielectric barrier discharge to occur. With the inert gas flowing through the cylindrical dielectric tube 611, when the high AC excitation voltage power source 615 is energized, a low-frequency high AC voltage is applied between the high-voltage electrode 612 and each of the upper and lower ground electrodes 613 and 614 located above and below. Consequently, an electric discharge occurs within the area sandwiched between the two ground electrodes 613 and 614. This electric discharge is induced through the dielectric coating layer (the wall surface of the cylindrical dielectric tube 611), and therefore, is a form of dielectric barrier discharge, whereby the plasma generation gas flowing through the cylindrical dielectric tube 611 is ionized over a wide area, forming a cloud of plasma (atmospheric-pressure non-equilibrium plasma).
The two ground electrodes 613 and 614 arranged so as to sandwich the high-voltage electrode 612 in between prevents the plasma generated by the electric discharge from spreading into the upstream and downstream sections of the cylindrical dielectric tube 611, whereby the effective plasma generation area is confined to the space between the two ground electrodes 613 and 614.
In the BID, the dielectric material which covers the surface of the plasma generation electrodes in the previously described manner prevents an emission of thermions or secondary electrons from the surface of the metallic electrodes. Furthermore, since the plasma generated by the dielectric barrier discharge is a non-equilibrium plasma with low-temperature neutral gas, various factors which cause a fluctuation of the plasma are suppressed, such as a temperature fluctuation in the discharging section or an emission of gas from the inner wall of the quartz tube due to the heat. As a result, the BID can maintain plasma in a stable form and thereby achieve a higher level of signal-to-noise (SN) ratio than the flame ionization detector (FID), which is the most commonly used type of detector for GC.
In general, there are two types of “dielectric barrier discharge”: an electric discharge generated by a configuration in which only one of the high-voltage and ground electrodes is covered with a dielectric body (which is hereinafter called the “single-side barrier discharge”); and an electric discharge generated by a configuration in which both of the high-voltage and ground electrodes are covered with a dielectric body (which is hereinafter called the “double-side barrier discharge”. Non Patent Literature 1 discloses the result of a study in which two discharging sections respectively employing those two configurations were constructed and their detector outputs in a BID-equivalent structure were compared, which demonstrated that a higher SN ratio could be achieved with the double-side barrier discharge than with the single-side barrier discharge.
As the inert gas for plasma generation in such a BID, helium (He) gas or argon (Ar) gas (or He gas with a trace amount of Ar gas added) is particularly widely used in practice. The reasons for using those gases are as follows:
(1) He gas: The discharge light generated by using He gas has an extremely high energy level of approximately 17.7 eV, making it possible to ionize and detect the atoms and molecules of most substances except for neon (Ne) and He. This is particularly useful for the detection of inorganic substances, since FIDs cannot ionize (and therefore cannot detect) inorganic substances.
(2) Ar gas (or He gas with a trace amount of Ar gas added): The energy level of discharge light generated by using Ar gas is approximately 11.7 eV and cannot ionize inorganic substances, as with the FID. This characteristic is useful in the case of specifically detecting organic substances. For example, in the case of detecting a trace amount of organic substance in an aqueous solution, the trace amount of organic substance of interest can be easily detected since the water used as the solvent cannot be detected.
Since the discharge characteristics vary depending on the kind of gas, the optimum electrode arrangement (e.g. the width of each electrode and the spacing of the electrodes) in the discharging section of the BID also changes depending on whether He gas or Ar gas is used as the inert gas. Accordingly, BIDs are configured to allow users to prepare a plurality of cylindrical dielectric tubes with different electrode arrangements and select a cylindrical dielectric tube having a suitable electrode arrangement for the kind of gas to be used. In the following description, a BID which uses Ar gas (or He gas with a trace amount of Ar gas added) as the plasma generation gas is called the “Ar-BID”. Similarly, a BID which uses lie gas as the plasma generation gas is called the “He-BID”.
FIG. 9 is a graph obtained by plotting the discharge initiation voltage for He and Ar at atmospheric pressure against the inter-electrode distance based on Paschen's law which is an empirical law concerning the discharge voltage for spark discharge. As can be seen in the graph, when the inter-electrode distance is the same, the discharge initiation voltage for Ar is approximately two times as high as the discharge initiation voltage for lie. In other words, provided that the device should be operated at the same discharge initiation voltage, the inter-electrode distance for Ar needs to be equal to or shorter than one half of the distance for He. Since there are also other parameters affecting the dielectric barrier discharge employed in BIDs, such as the material of the dielectric body, gas purity, frequency of the discharge power source, and waveform of the power source, it is difficult to predict an optimum electrode arrangement and discharging conditions from Paschen's law which is the empirical law concerning spark discharge. However, from the foregoing discussion, it is at least possible to conclude that the Ar-BID requires a shorter inter-electrode distance between the plasma generation electrodes (or a higher discharge voltage) than the He-BID.